Establishment of real-time monitoring technique for blood viscosity during extracorporeal circulation treatment is needed. We have demonstrated that red blood cell (RBC, hereafter) aggregation degree, which highly correlates with the blood viscosity, can be estimated by the ultrasound spectroscopic observation of the RBC and the defining of the peak frequency by the Fast Fourier Transform process. In this study, the peak frequency was measured to investigate the aggregation degree of RBC flowing in a simple heart-lung machine. In addition, to disassemble the RBC aggregations that might cause the clogging of the circuit, a disassemble technique of the RBC aggregations using a focused 5 MHz ultrasonic transducer was also proposed. As the first step of this study, the disassembling of the RBC aggregations was carried out under the still condition and its effect was confirmed by the peak frequency method.

This paper proposes a sleep visualization system through indirect biofeedback for patients' behavioral changes and improves sleep quality. In a super-aging society that has been developing in the world, efficient and high-quality cares for elderly people are needed. Especially in nursing care, a lot of patients have sleep problems due to reduced stamina and activity, and it eventually results in the increase of caregivers' strain and burden. For solving this problem, a sleep visualization system using indirect biofeedback is developed for patients aiming to behavioral changes and improving the sleep quality. To validate the proposed system, an experiment was conducted with five elderly persons. As a result, it is verified that some patients were interested in their sleep quality through the proposed system and the sleep score could be improved.

Speech separation refers to the separation of mixed utterances in which multiple people are speaking simultaneously. This topic was created to address the cocktail party problem, and it has also evolved into a front-end procedure for speech-related research recently. There are two types of models for addressing it: classification methods and regression methods. Classification methods avoid considering the permutation problem, and regression methods are more precise. Deep clustering (DC) is a common deep learning classification technique that uses a deep embedding network to embed signals in the underlying manifold and allows signals with similar properties to cluster together in the embedded space. The ideal binary mask (IBM) supervises the model. Even if it is entirely isolated via supervised binary masking, this setting is susceptible to mistake. To remedy this flaw, this paper proposes a network with two modules based on deep embedding. More precise masks, such as the ideal ratio mask (IRM), can now be used instead of IBM. Furthermore, the cascade structure may preserve DC's high performance. On average, the suggested model beats the original DC model by 1.55dB in SNR, 4.45dB in SDR, 4.41dB in SDRi, 0.16 in STOI, and 0.3 in PESQ on a mixture of Japanese Newspaper Article Sentences (JNAS). Finally, a comparison is made with ChimeraNet which has a similar structure, the results prove the proposed model's superiority over previous DC-related studies.

Recently, unsupervised learning-based approaches for optical flow estimation have been actively researched. In unsupervised settings, the difference between the input image and the reconstructed image created from the estimated optical flow is minimized to learn the optical flow. Conventional learning methods mainly treated the difference as the pixel-by-pixel brightness error, which might lead to decreasing accuracy of the optical flow because the learning-strategy cannot take the textures into account sufficiently. To deal with the problem, in addition to the brightness error, we propose the introduction of adversarial learning into the evaluation of the input image and the reconstructed image. Our main contribution is that we develop a learning-strategy to capture the textures and the proposed method outperforms the conventional methods on the KITTI benchmarks.

Targeting that accurately captures customer needs is expected to become even more important in the future in order to overcome competitors. Some beauty salon customers also come from local areas to urban stores. Even in downtowns, customer demographics and customer preferences vary greatly depending on the region. Therefore, it can be inferred that there is some relationship behind the location of the store and the needs of customers. The purpose of this research is to understand the characteristics of customer needs and the relationship between regions. In this paper, we analyze the difference in the customer segment between regional areas in the beauty salon industry, through analysis of customer review data posted on the beauty salon reservation site. By applying non-negative matrix factorization method to the customer reviews for the stores in the commercial area to be analyzed, we extracted the topics related to the needs in each region. Furthermore, we attempted to visualize the needs caused by the commercial area by associating the response magnitude of this topic with the location information of the store where the review text was posted.

This paper investigates the Neimark-Sacker bifurcation points derivation method using the nested-layer particle swarm optimization (NLPSO) in nonlinear dynamical systems. In nonlinear dynamical systems, Neimark-Sacker bifurcation occurs when a complex conjugate characteristic multiplier crosses the unit circle by changing the parameters. NLPSO is not only under no careful initial values and complex hand calculation but also the advantage of deriving bifurcation points regardless of the stability of the systems. In order to obtain the Neimark-Sacker bifurcation parameters by NLPSO, the characteristic multiplier of the objective function must be set to a complex number. This method derives the local bifurcation points such as Neimark-Sacker bifurcation, period-doubling bifurcation, and saddle-node bifurcation by changing only the declination of the characteristic multiplier without changing the objective function and algorithm. As an application example, we give the results of deriving Neimark-Sacker bifurcation points in two-dimensional discrete dynamical and non-autonomous systems.

Cloud computing is being increasingly employed in the development of systems. For such systems, rapid cost estimation is required to start new business. This paper proposes a “Pay-Per-Use” billing model, and providers offer various strategic pricing options. There are many cost structures, and hence, users need to consider many pricing options when estimating system costs. Although companies often want to estimate the cost of their systems, it is difficult to obtain a comprehensive estimate that considers all components. Previously, researchers have implemented simulation-based and analytical approaches to solve this problem. In this study, we develop a cost-estimation method that employs directed acyclic graph (DAG)-based representation and matrix operations. Our method emphasizes simplicity through the use of a systematic procedure that can estimate the costs of new services in the pre-design stage.

The authors have proposed a combinatorial optimization method based on hierarchical structure in solution space which method performs for several types of combinatorial optimization problems at a high level. In this letter, we indicate an issue which occurs deterioration of search efficiency and propose a method which can solve the issue. The usefulness of the developed method is confirmed by numerical experiments using benchmark problems.

Consensus building is a hard task because the members have respective preferences and they have sometime conflicting interests. Multi-person menu planning is such a consensus building task. In this paper, we propose an island style interactive evolutionary computation with constraint agents. The agents implicitly impose constraints without disturbing consensus building. Our method for menu planning has been supported in two experiments.